IMPACT OF AFRICAN MEDIEVAL MIGRATIONS IN SPAIN
Fig. 1. Frequency distribution of mtDNA haplogroups in South Iberian Peninsula, Northwest Africa as well as Medieval (MP) and present (PP) Priego de Co´rdoba. The star indicates the geographical location of Priego de Co´rdoba in Spain. Left pie graphs represent the relative frequencies of Eurasiatic (white), Nor-African (dotted), sub-Saharan (black), and African L3 (grey) haplogroups. Right pie graphs represent the relative frequencies of African haplogroups in each sample, with colors as in left charts.
Fig. 2. L1b network constructed with the European haplo- types carrying the 16175 transition. Star represents the basic HVRI L1b motif (126 187 189 223 264 270 278 311). Numbers along links refer to nucleotide positions (mentioned in Anderson et al. 1981) minus 16,000. GER, Germany; RUS, Russian; MP, Medieval Priego.
congruent with this proposed African pre-Muslim arrival to Europe.
Iberia. The unique sharing of L1b (126 127 189 223 264 278 311) with the Sahara points to this area as the most probable origin. Nevertheless, the high number of non- shared lineages impedes the determination of the precise African origin. Moreover, the fact that the closest sequences to the two L1b haplotypes with the 16175 transition are not in Africa but in Germany (Richards et al., 2000) and Russia (Malyarchuk et al., 2004) clouds the origin of these haplotypes. It could be simply that they are also present in Africa but not yet detected. However, when we revised published and our unpub- lished HVRI data from Europe (11,511) and Africa (4,566), we collected 31 L1b European haplotypes, four of which are 16175 carriers, giving a frequency of 0.129 for this type in Europe. On the contrary, from 310 L1b Afri- cans none of them has the 16175 transition. Supposing that Africa has the same frequency as in Europe, the bi- nomial probability of not finding this motif in the African sample would be 2.5 3 1019. Therefore, the most proba- ble situation is that the ancestor of this motif (Fig. 2) arrived from Africa to Europe where the 16175 mutation occurred. The divergence time estimation for this clade in Europe is around 20,180 6 16,144 years, pointing to a prehistoric arrival, in Europe, of the basic African motif. It could also be that L1b with the 16175 transition was carried to the Iberian Peninsula with the Muslim invaders and that this motif was lost in Africa. However, there are not important NW African demographic move- ments registered in recent times that could explain this loss. Furthermore, this clade has a relatively high diver- sity in Europe (three different haplotypes from four indi- viduals); therefore, if these types had arrived from Africa in recent times, we would have to suppose that this great diversity should already be in Africa. Moreover, the presence of a North African M1 representative, recently detected, in historic Basque remains that pre- date the Muslim invasion (Alzualde et al., in press) is
Regarding those L3 African lineages only found in cur- rent Priego and NW Africa (belonging to L3h and L3e5 haplogroups), their limited distribution favors a more recent arrival to Iberia, in this case from Tunisia, as both lineages have only been found in Berber popula- tions from this country.
The results of this research have shown how popula- tion studies on ancient DNA can contribute to a better understanding of the origin of present genetic composi- tion of human populations. Studies combining genetic data from different periods should be a new objective in human evolutionary studies, as they may help to solve some of the controversies surrounding the arrival of spe- cific markers, and the impact of different human migra- tions on the present genetic structure of the populations.
The medieval Priego sample showed greater affinities to North-Africa than other Iberian Peninsula samples including that of present day Priego. Haplotype analysis revealed that some African haplotypes detected in medi- eval Priego have matches with samples of precise north- African origin as Tunisia, west-Sahara or the Canary Islands pointing to well documented historic connections with this area. However, medieval Priego L1b lineages carrying the 16175 transition have their most related counterparts in Europe instead of Africa. The coales- cence age for these L1b lineages is compatible with a minor prehistoric African influence on Priego that also reached other European areas.
We thank the valuable scientific assistance and sup- port of Vicente M. Cabrera in carrying out this research. We are grateful to Emmanuel Cleuvenot and Rafael Car- mona for the scientific contribution to the anthropologi-
American Journal of Physical Anthropology—DOI 10.1002/ajpa